[OpenID board] May 19, 2010 OpenID Board Meeting Minutes

Chris Messina chris.messina at gmail.com
Tue Jun 1 02:34:04 UTC 2010

At the very least, the 2.x series of changes should investigate the net-net
adoption of all of the features of OpenID and prioritize and de-prioritize

Supporting XRIs have been reported to add complexity to the discovery code,
and further, seem to have little adoption in the mainstream and few
implementations in the wild. I'd be eager to hear specific stats that
contradict that sentiment, but I don't want to hold on to features purely
out of nostalgia.

This points to yet another reason why I worry about the v.Next naming: if we
can't even cut, cut, cut from the 2.0 spec to create a leaner, simpler
protocol that's easier to implement and support, I have a hard time
imagining how we're going to arrive at a simpler, stream-lined technology
when v.Next sounds like it's chartered to include everything and the kitchen

Perhaps along with the MRD/PRD that Brian Kissel wants, we should also
produce a DRD — a developer requirements document — that provides insight
into which features of OpenID have actually been implemented by the most
successful OPs (based on market adoption and usage). That is, in order to be
taken into consideration for this requirements document, you have to have
already deployed a public OpenID Provider and have people using it.

It seems that that would be a useful filter to gather real-world feedback
that can lead us forward for 2.x.


On Mon, May 31, 2010 at 7:05 PM, John Bradley <jbradley at mac.com> wrote:

> Dick,
> You know that XRI is used by multiple OP.   It is not something a RP site
> demands.
> In discovery going forward we need to look at how and if XRI identifiers
> are supported.
> Because LRDD and XRI are so close you can normalize a XRI by adding @
> xri.net to the end and doing webfinger/LRDD resolution to get a XRD.
> People may be opposed to XRI on philosophical grounds, and want it removed.
> Breaking peoples openID without a good reason should be avoided.
> I prefer to look at it as coming up with a suitable discovery flow and then
> determining what identifier formats are appropriate.
> If at the end of the day XRI can't be supported we will discover that.   I
> would prefer not to make that a precondition.
> There are a number of equally under-utilized features of AX we could look
> at:)
> John B.
> On 2010-05-31, at 9:48 PM, Dick Hardt wrote:
> >
> > On 2010-05-31, at 6:33 PM, Nat Sakimura wrote:
> >
> >> I do not understand why we need to remove XRI.
> >> XRI being complex etc. is an illusion or badly written code.
> >> If XRI is complex, acct: URI is complex, too.
> >> They virtually are the same thing from the processing point of view.
> >> Keeping the identifier compatibility is really important for the RPs and
> Users.
> >
> > I don't think there is much usage of URI, and many sites are not
> supporting it. Removing under utilized features that are not adding value
> simplifies the specs and implementations.
> >
> > Nat: do you know of sites that are heavily using XRIs?
> >
> > -- Dick
> > _______________________________________________
> > board mailing list
> > board at lists.openid.net
> > http://lists.openid.net/mailman/listinfo/openid-board
> _______________________________________________
> board mailing list
> board at lists.openid.net
> http://lists.openid.net/mailman/listinfo/openid-board

Chris Messina
Open Web Advocate, Google

Personal: http://factoryjoe.com
Follow me on Buzz: http://buzz.google.com/chrismessina
...or Twitter: http://twitter.com/chrismessina

This email is:   [ ] shareable    [X] ask first   [ ] private
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openid.net/pipermail/openid-board/attachments/20100531/87f0a1a1/attachment-0003.html>

More information about the board mailing list